[CalendarServer-changes] [15691] CalendarServer/trunk/contrib/performance/loadtest/standard-configs
source_changes at macosforge.org
source_changes at macosforge.org
Wed Jun 22 17:42:03 PDT 2016
Revision: 15691
http://trac.calendarserver.org//changeset/15691
Author: sagen at apple.com
Date: 2016-06-22 17:42:02 -0700 (Wed, 22 Jun 2016)
Log Message:
-----------
Remove obsolete sim plists and add a new one which does consistent invites
Added Paths:
-----------
CalendarServer/trunk/contrib/performance/loadtest/standard-configs/constant-invites.plist
Removed Paths:
-------------
CalendarServer/trunk/contrib/performance/loadtest/standard-configs/event-updates-only.plist
CalendarServer/trunk/contrib/performance/loadtest/standard-configs/events-only.plist
CalendarServer/trunk/contrib/performance/loadtest/standard-configs/invites-accepts.plist
CalendarServer/trunk/contrib/performance/loadtest/standard-configs/invites-only-recurring.plist
CalendarServer/trunk/contrib/performance/loadtest/standard-configs/invites-only.plist
Added: CalendarServer/trunk/contrib/performance/loadtest/standard-configs/constant-invites.plist
===================================================================
--- CalendarServer/trunk/contrib/performance/loadtest/standard-configs/constant-invites.plist (rev 0)
+++ CalendarServer/trunk/contrib/performance/loadtest/standard-configs/constant-invites.plist 2016-06-23 00:42:02 UTC (rev 15691)
@@ -0,0 +1,340 @@
+<?xml version="1.0" encoding="UTF-8"?>
+
+<!--
+ Copyright (c) 2011-2016 Apple Inc. All rights reserved.
+
+ Licensed under the Apache License, Version 2.0 (the "License");
+ you may not use this file except in compliance with the License.
+ You may obtain a copy of the License at
+
+ http://www.apache.org/licenses/LICENSE-2.0
+
+ Unless required by applicable law or agreed to in writing, software
+ distributed under the License is distributed on an "AS IS" BASIS,
+ WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ See the License for the specific language governing permissions and
+ limitations under the License.
+ -->
+
+<!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">
+<plist version="1.0">
+ <dict>
+ <!-- Define the kinds of software and user behavior the load simulation
+ will simulate. -->
+ <key>clients</key>
+
+ <!-- Have as many different kinds of software and user behavior configurations
+ as you want. Each is a dict -->
+ <array>
+
+ <dict>
+
+ <!-- Here is a OS X client simulator. -->
+ <key>software</key>
+ <string>contrib.performance.loadtest.ical.OS_X_10_11</string>
+
+ <!-- Arguments to use to initialize the OS_X_10_11 instance. -->
+ <key>params</key>
+ <dict>
+ <!-- Name that appears in logs. -->
+ <key>title</key>
+ <string>main</string>
+
+ <!-- OS_X_10_11 can poll the calendar home at some interval. This is
+ in seconds. -->
+ <key>calendarHomePollInterval</key>
+ <integer>30</integer>
+
+ <!-- If the server advertises xmpp push, OS_X_10_7 can wait for notifications
+ about calendar home changes instead of polling for them periodically. If
+ this option is true, then look for the server advertisement for xmpp push
+ and use it if possible. Still fall back to polling if there is no xmpp push
+ advertised. -->
+ <key>supportPush</key>
+ <true/>
+
+ <key>supportAmpPush</key>
+ <true/>
+
+ <key>unauthenticatedPercentage</key>
+ <integer>40</integer>
+
+ </dict>
+
+ <!-- The profiles define certain types of user behavior on top of the
+ client software being simulated. -->
+ <key>profiles</key>
+ <array>
+
+ <!-- This profile invites some number of new attendees to new events. -->
+ <dict>
+ <key>class</key>
+ <string>contrib.performance.loadtest.profiles.Inviter</string>
+
+ <key>params</key>
+ <dict>
+ <key>enabled</key>
+ <true/>
+
+ <!-- Define the frequency at which new invitations will be sent out. -->
+ <key>sendInvitationDistribution</key>
+ <dict>
+ <key>type</key>
+ <string>contrib.performance.stats.FixedDistribution</string>
+ <key>params</key>
+ <dict>
+ <!-- seconds -->
+ <key>value</key>
+ <integer>60</integer>
+ </dict>
+ </dict>
+
+ <!-- Define the distribution of who will be invited to an event.
+
+ When inviteeClumping is turned on each invitee is based on a sample of
+ users "close to" the organizer based on account index. If the clumping
+ is too "tight" for the requested number of attendees, then invites for
+ those larger numbers will simply fail (the sim will report that situation).
+
+ When inviteeClumping is off invitees will be sampled across an entire
+ range of account indexes. In this case the distribution ought to be a
+ UniformIntegerDistribution with min=0 and max set to the number of accounts.
+ -->
+ <key>inviteeDistribution</key>
+ <dict>
+ <key>type</key>
+ <string>contrib.performance.stats.UniformIntegerDistribution</string>
+ <key>params</key>
+ <dict>
+ <!-- The minimum value (inclusive) of the uniform distribution. -->
+ <key>min</key>
+ <integer>0</integer>
+ <!-- The maximum value (exclusive) of the uniform distribution. -->
+ <key>max</key>
+ <integer>99</integer>
+ </dict>
+ </dict>
+
+ <key>inviteeClumping</key>
+ <true/>
+
+ <!-- Define the distribution of how many attendees will be invited to an event.
+
+ LogNormal is the best fit to observed data.
+
+
+ For LogNormal "mode" is the peak, "mean" is the mean value. For invites,
+ mode should typically be 1, and mean whatever matches the user behavior.
+ Our typical mean is 6.
+ -->
+ <key>inviteeCountDistribution</key>
+ <dict>
+ <key>type</key>
+ <string>contrib.performance.stats.FixedDistribution</string>
+ <key>params</key>
+ <dict>
+ <!-- count -->
+ <key>value</key>
+ <integer>6</integer>
+ </dict>
+ </dict>
+
+ <!-- Percentage of time to do a lookup for invitee -->
+ <key>inviteeLookupPercentage</key>
+ <integer>0</integer>
+
+ <!-- Define how start times (DTSTART) for the randomly generated events
+ will be selected. This is an example of a "Distribution" parameter. The value
+ for most "Distribution" parameters are interchangeable and extensible. -->
+ <key>eventStartDistribution</key>
+ <dict>
+
+ <!-- This distribution is pretty specialized. It produces timestamps
+ in the near future, limited to certain days of the week and certain hours
+ of the day. -->
+ <key>type</key>
+ <string>contrib.performance.stats.WorkDistribution</string>
+
+ <key>params</key>
+ <dict>
+ <!-- These are the days of the week the distribution will use. -->
+ <key>daysOfWeek</key>
+ <array>
+ <string>mon</string>
+ <string>tue</string>
+ <string>wed</string>
+ <string>thu</string>
+ <string>fri</string>
+ </array>
+
+ <!-- The earliest hour of a day at which an event might be scheduled. -->
+ <key>beginHour</key>
+ <integer>8</integer>
+
+ <!-- And the latest hour of a day (at which an event will be scheduled
+ to begin!). -->
+ <key>endHour</key>
+ <integer>16</integer>
+
+ <!-- The timezone in which the event is scheduled. (XXX Does this
+ really work right?) -->
+ <key>tzname</key>
+ <string>America/Los_Angeles</string>
+ </dict>
+ </dict>
+
+ <!-- Define how recurrences are created. -->
+ <key>recurrenceDistribution</key>
+ <dict>
+
+ <!-- This distribution is pretty specialized. We have a fixed set of
+ RRULEs defined for this distribution and pick each based on a
+ weight. -->
+ <key>type</key>
+ <string>contrib.performance.stats.RecurrenceDistribution</string>
+
+ <key>params</key>
+ <dict>
+ <!-- False to disable RRULEs -->
+ <key>allowRecurrence</key>
+ <false/>
+
+ <!-- These are the weights for the specific set of RRULEs. -->
+ <key>weights</key>
+ <dict>
+ </dict>
+ </dict>
+ </dict>
+
+ <!-- Percentage of time to attach a file -->
+ <key>fileAttachPercentage</key>
+ <integer>0</integer>
+
+ <!-- Define the attachment size distribution. -->
+ <key>fileSizeDistribution</key>
+ <dict>
+ <key>type</key>
+ <string>contrib.performance.stats.NormalDistribution</string>
+ <key>params</key>
+ <dict>
+ <!-- mu gives the mean of the normal distribution (in seconds). -->
+ <key>mu</key>
+ <integer>500000</integer>
+
+ <!-- and sigma gives its standard deviation. -->
+ <key>sigma</key>
+ <integer>100000</integer>
+ </dict>
+ </dict>
+
+ </dict>
+ </dict>
+
+ <!-- This profile accepts invitations to events, handles cancels, and
+ handles replies received. -->
+ <dict>
+ <key>class</key>
+ <string>contrib.performance.loadtest.profiles.Accepter</string>
+
+ <key>params</key>
+ <dict>
+ <key>enabled</key>
+ <true/>
+
+ <!-- Define how long to wait after seeing a new invitation before
+ accepting it.
+ -->
+ <key>acceptDelayDistribution</key>
+ <dict>
+ <key>type</key>
+ <string>contrib.performance.stats.FixedDistribution</string>
+ <key>params</key>
+ <dict>
+ <!-- seconds-->
+ <key>value</key>
+ <integer>30</integer>
+ </dict>
+ </dict>
+ </dict>
+ </dict>
+
+ <!-- Removes events from calendars exceeding a threshold -->
+ <dict>
+ <key>class</key>
+ <string>contrib.performance.loadtest.profiles.EventCountLimiter</string>
+
+ <key>params</key>
+ <dict>
+ <key>enabled</key>
+ <true/>
+
+ <!-- Define the interval (in seconds) at which this profile will check
+ for too-large collections. -->
+ <key>interval</key>
+ <integer>60</integer>
+
+ <!-- The upper bound. -->
+ <key>eventCountLimit</key>
+ <integer>100</integer>
+
+ </dict>
+ </dict>
+
+
+ </array>
+
+ <!-- Determine the frequency at which this client configuration will
+ appear in the clients which are created by the load tester. -->
+ <key>weight</key>
+ <integer>1</integer>
+ </dict>
+
+ <dict>
+
+ <!-- Here is a more passive OS X client simulator. -->
+ <key>software</key>
+ <string>contrib.performance.loadtest.ical.OS_X_10_11</string>
+
+ <!-- Arguments to use to initialize the OS_X_10_11 instance. -->
+ <key>params</key>
+ <dict>
+ <!-- Name that appears in logs. -->
+ <key>title</key>
+ <string>idle</string>
+
+ <!-- OS_X_10_11 can poll the calendar home at some interval. This is
+ in seconds. -->
+ <key>calendarHomePollInterval</key>
+ <integer>30</integer>
+
+ <!-- If the server advertises xmpp push, OS_X_10_7 can wait for notifications
+ about calendar home changes instead of polling for them periodically. If
+ this option is true, then look for the server advertisement for xmpp push
+ and use it if possible. Still fall back to polling if there is no xmpp push
+ advertised. -->
+ <key>supportPush</key>
+ <true/>
+
+ <key>supportAmpPush</key>
+ <true/>
+
+ <key>unauthenticatedPercentage</key>
+ <integer>40</integer>
+
+ </dict>
+
+ <!-- The profiles define certain types of user behavior on top of the
+ client software being simulated. -->
+ <key>profiles</key>
+ <array>
+
+ </array>
+ <!-- Determine the frequency at which this client configuration will
+ appear in the clients which are created by the load tester. -->
+ <key>weight</key>
+ <integer>2</integer>
+ </dict>
+
+ </array>
+ </dict>
+</plist>
Deleted: CalendarServer/trunk/contrib/performance/loadtest/standard-configs/event-updates-only.plist
===================================================================
--- CalendarServer/trunk/contrib/performance/loadtest/standard-configs/event-updates-only.plist 2016-06-22 21:47:05 UTC (rev 15690)
+++ CalendarServer/trunk/contrib/performance/loadtest/standard-configs/event-updates-only.plist 2016-06-23 00:42:02 UTC (rev 15691)
@@ -1,543 +0,0 @@
-<?xml version="1.0" encoding="UTF-8"?>
-
-<!--
- Copyright (c) 2011-2016 Apple Inc. All rights reserved.
-
- Licensed under the Apache License, Version 2.0 (the "License");
- you may not use this file except in compliance with the License.
- You may obtain a copy of the License at
-
- http://www.apache.org/licenses/LICENSE-2.0
-
- Unless required by applicable law or agreed to in writing, software
- distributed under the License is distributed on an "AS IS" BASIS,
- WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- See the License for the specific language governing permissions and
- limitations under the License.
- -->
-
-<!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">
-<plist version="1.0">
- <dict>
- <!-- Define the kinds of software and user behavior the load simulation
- will simulate. -->
- <key>clients</key>
-
- <!-- Have as many different kinds of software and user behavior configurations
- as you want. Each is a dict -->
- <array>
-
- <dict>
-
- <!-- Here is a Lion iCal simulator. -->
- <key>software</key>
- <string>contrib.performance.loadtest.ical.OS_X_10_7</string>
-
- <!-- Arguments to use to initialize the client instance. -->
- <key>params</key>
- <dict>
- <!-- Name that appears in logs. -->
- <key>title</key>
- <string>10.7</string>
-
- <!-- Client can poll the calendar home at some interval. This is
- in seconds. -->
- <key>calendarHomePollInterval</key>
- <integer>300000</integer>
-
- <!-- If the server advertises xmpp push, OS X 10.6 can wait for notifications
- about calendar home changes instead of polling for them periodically. If
- this option is true, then look for the server advertisement for xmpp push
- and use it if possible. Still fall back to polling if there is no xmpp push
- advertised. -->
- <key>supportPush</key>
- <false />
- <key>supportAmpPush</key>
- <false />
- </dict>
-
- <!-- The profiles define certain types of user behavior on top of the
- client software being simulated. -->
- <key>profiles</key>
- <array>
-
- <!-- First an event-creating profile, which will periodically create
- new events at a random time on a random calendar. -->
- <dict>
- <key>class</key>
- <string>contrib.performance.loadtest.profiles.Eventer</string>
-
- <key>params</key>
- <dict>
- <key>enabled</key>
- <false/>
-
- <!-- Define the interval (in seconds) at which this profile will use
- its client to create a new event. -->
- <key>interval</key>
- <integer>20</integer>
-
- <!-- Define how start times (DTSTART) for the randomly generated events
- will be selected. This is an example of a "Distribution" parameter. The value
- for most "Distribution" parameters are interchangeable and extensible. -->
- <key>eventStartDistribution</key>
- <dict>
-
- <!-- This distribution is pretty specialized. It produces timestamps
- in the near future, limited to certain days of the week and certain hours
- of the day. -->
- <key>type</key>
- <string>contrib.performance.stats.WorkDistribution</string>
-
- <key>params</key>
- <dict>
- <!-- These are the days of the week the distribution will use. -->
- <key>daysOfWeek</key>
- <array>
- <string>mon</string>
- <string>tue</string>
- <string>wed</string>
- <string>thu</string>
- <string>fri</string>
- </array>
-
- <!-- The earliest hour of a day at which an event might be scheduled. -->
- <key>beginHour</key>
- <integer>8</integer>
-
- <!-- And the latest hour of a day (at which an event will be scheduled
- to begin!). -->
- <key>endHour</key>
- <integer>16</integer>
-
- <!-- The timezone in which the event is scheduled. (XXX Does this
- really work right?) -->
- <key>tzname</key>
- <string>America/Los_Angeles</string>
- </dict>
- </dict>
-
- <!-- Define how recurrences are created. -->
- <key>recurrenceDistribution</key>
- <dict>
-
- <!-- This distribution is pretty specialized. We have a fixed set of
- RRULEs defined for this distribution and pick each based on a
- weight. -->
- <key>type</key>
- <string>contrib.performance.stats.RecurrenceDistribution</string>
-
- <key>params</key>
- <dict>
- <!-- False to disable RRULEs -->
- <key>allowRecurrence</key>
- <false/>
-
- <!-- These are the weights for the specific set of RRULEs. -->
- <key>weights</key>
- <dict>
- <!-- Half of all events will be non-recurring -->
- <key>none</key>
- <integer>50</integer>
-
- <!-- Daily and weekly are pretty common -->
- <key>daily</key>
- <integer>10</integer>
- <key>weekly</key>
- <integer>20</integer>
-
- <!-- Monthly, yearly, daily & weekly limit not so common -->
- <key>monthly</key>
- <integer>2</integer>
- <key>yearly</key>
- <integer>1</integer>
- <key>dailylimit</key>
- <integer>2</integer>
- <key>weeklylimit</key>
- <integer>5</integer>
-
- <!-- Work days pretty common -->
- <key>workdays</key>
- <integer>10</integer>
- </dict>
- </dict>
- </dict>
- </dict>
- </dict>
-
- <!-- This profile will create a new event, and then periodically update the ACKNOWLEDGED property. -->
- <dict>
- <key>class</key>
- <string>contrib.performance.loadtest.profiles.EventUpdater</string>
-
- <key>params</key>
- <dict>
- <key>enabled</key>
- <true/>
-
- <!-- Define the interval (in seconds) at which this profile will use
- its client to create a new event. -->
- <key>interval</key>
- <integer>5</integer>
-
- <!-- Define how start times (DTSTART) for the randomly generated events
- will be selected. This is an example of a "Distribution" parameter. The value
- for most "Distribution" parameters are interchangeable and extensible. -->
- <key>eventStartDistribution</key>
- <dict>
-
- <!-- This distribution is pretty specialized. It produces timestamps
- in the near future, limited to certain days of the week and certain hours
- of the day. -->
- <key>type</key>
- <string>contrib.performance.stats.WorkDistribution</string>
-
- <key>params</key>
- <dict>
- <!-- These are the days of the week the distribution will use. -->
- <key>daysOfWeek</key>
- <array>
- <string>mon</string>
- <string>tue</string>
- <string>wed</string>
- <string>thu</string>
- <string>fri</string>
- </array>
-
- <!-- The earliest hour of a day at which an event might be scheduled. -->
- <key>beginHour</key>
- <integer>8</integer>
-
- <!-- And the latest hour of a day (at which an event will be scheduled
- to begin!). -->
- <key>endHour</key>
- <integer>16</integer>
-
- <!-- The timezone in which the event is scheduled. (XXX Does this
- really work right?) -->
- <key>tzname</key>
- <string>America/Los_Angeles</string>
- </dict>
- </dict>
-
- <!-- Define how recurrences are created. -->
- <key>recurrenceDistribution</key>
- <dict>
-
- <!-- This distribution is pretty specialized. We have a fixed set of
- RRULEs defined for this distribution and pick each based on a
- weight. -->
- <key>type</key>
- <string>contrib.performance.stats.RecurrenceDistribution</string>
-
- <key>params</key>
- <dict>
- <!-- False to disable RRULEs -->
- <key>allowRecurrence</key>
- <true/>
-
- <!-- These are the weights for the specific set of RRULEs. -->
- <key>weights</key>
- <dict>
- <!-- Half of all events will be non-recurring -->
- <key>none</key>
- <integer>50</integer>
-
- <!-- Daily and weekly are pretty common -->
- <key>daily</key>
- <integer>25</integer>
- <key>weekly</key>
- <integer>25</integer>
-
- <!-- Monthly, yearly, daily & weekly limit not so common -->
- <key>monthly</key>
- <integer>0</integer>
- <key>yearly</key>
- <integer>0</integer>
- <key>dailylimit</key>
- <integer>0</integer>
- <key>weeklylimit</key>
- <integer>0</integer>
-
- <!-- Work days pretty common -->
- <key>workdays</key>
- <integer>0</integer>
- </dict>
- </dict>
- </dict>
- </dict>
- </dict>
-
- <!-- This profile invites some number of new attendees to new events. -->
- <dict>
- <key>class</key>
- <string>contrib.performance.loadtest.profiles.RealisticInviter</string>
-
- <key>params</key>
- <dict>
- <key>enabled</key>
- <false/>
-
- <!-- Define the frequency at which new invitations will be sent out. -->
- <key>sendInvitationDistribution</key>
- <dict>
- <key>type</key>
- <string>contrib.performance.stats.NormalDistribution</string>
- <key>params</key>
- <dict>
- <!-- mu gives the mean of the normal distribution (in seconds). -->
- <key>mu</key>
- <integer>10</integer>
-
- <!-- and sigma gives its standard deviation. -->
- <key>sigma</key>
- <integer>5</integer>
- </dict>
- </dict>
-
- <!-- Define the distribution of who will be invited to an event.
-
- When inviteeClumping is turned on each invitee is based on a sample of
- users "close to" the organizer based on account index. If the clumping
- is too "tight" for the requested number of attendees, then invites for
- those larger numbers will simply fail (the sim will report that situation).
-
- When inviteeClumping is off invitees will be sampled across an entire
- range of account indexes. In this case the distribution ought to be a
- UniformIntegerDistribution with min=0 and max set to the number of accounts.
- -->
- <key>inviteeDistribution</key>
- <dict>
- <key>type</key>
- <string>contrib.performance.stats.UniformIntegerDistribution</string>
- <key>params</key>
- <dict>
- <!-- The minimum value (inclusive) of the uniform distribution. -->
- <key>min</key>
- <integer>0</integer>
- <!-- The maximum value (exclusive) of the uniform distribution. -->
- <key>max</key>
- <integer>99</integer>
- </dict>
- </dict>
-
- <key>inviteeClumping</key>
- <true/>
-
- <!-- Define the distribution of how many attendees will be invited to an event.
-
- LogNormal is the best fit to observed data.
-
-
- For LogNormal "mode" is the peak, "mean" is the mean value. For invites,
- mode should typically be 1, and mean whatever matches the user behavior.
- Our typical mean is 6.
- -->
- <key>inviteeCountDistribution</key>
- <dict>
- <key>type</key>
- <string>contrib.performance.stats.LogNormalDistribution</string>
- <key>params</key>
- <dict>
- <!-- mode - peak-->
- <key>mode</key>
- <integer>1</integer>
- <!-- mean - average-->
- <key>median</key>
- <integer>6</integer>
- <!-- maximum -->
- <key>maximum</key>
- <real>100</real>
- </dict>
- </dict>
-
- <!-- Define how start times (DTSTART) for the randomly generated events
- will be selected. This is an example of a "Distribution" parameter. The value
- for most "Distribution" parameters are interchangeable and extensible. -->
- <key>eventStartDistribution</key>
- <dict>
-
- <!-- This distribution is pretty specialized. It produces timestamps
- in the near future, limited to certain days of the week and certain hours
- of the day. -->
- <key>type</key>
- <string>contrib.performance.stats.WorkDistribution</string>
-
- <key>params</key>
- <dict>
- <!-- These are the days of the week the distribution will use. -->
- <key>daysOfWeek</key>
- <array>
- <string>mon</string>
- <string>tue</string>
- <string>wed</string>
- <string>thu</string>
- <string>fri</string>
- </array>
-
- <!-- The earliest hour of a day at which an event might be scheduled. -->
- <key>beginHour</key>
- <integer>8</integer>
-
- <!-- And the latest hour of a day (at which an event will be scheduled
- to begin!). -->
- <key>endHour</key>
- <integer>16</integer>
-
- <!-- The timezone in which the event is scheduled. (XXX Does this
- really work right?) -->
- <key>tzname</key>
- <string>America/Los_Angeles</string>
- </dict>
- </dict>
-
- <!-- Define how recurrences are created. -->
- <key>recurrenceDistribution</key>
- <dict>
-
- <!-- This distribution is pretty specialized. We have a fixed set of
- RRULEs defined for this distribution and pick each based on a
- weight. -->
- <key>type</key>
- <string>contrib.performance.stats.RecurrenceDistribution</string>
-
- <key>params</key>
- <dict>
- <!-- False to disable RRULEs -->
- <key>allowRecurrence</key>
- <true/>
-
- <!-- These are the weights for the specific set of RRULEs. -->
- <key>weights</key>
- <dict>
- <!-- Half of all events will be non-recurring -->
- <key>none</key>
- <integer>50</integer>
-
- <!-- Daily and weekly are pretty common -->
- <key>daily</key>
- <integer>10</integer>
- <key>weekly</key>
- <integer>20</integer>
-
- <!-- Monthly, yearly, daily & weekly limit not so common -->
- <key>monthly</key>
- <integer>2</integer>
- <key>yearly</key>
- <integer>1</integer>
- <key>dailylimit</key>
- <integer>2</integer>
- <key>weeklylimit</key>
- <integer>5</integer>
-
- <!-- Work days pretty common -->
- <key>workdays</key>
- <integer>10</integer>
- </dict>
- </dict>
- </dict>
- </dict>
- </dict>
-
- <!-- This profile accepts invitations to events, handles cancels, and
- handles replies received. -->
- <dict>
- <key>class</key>
- <string>contrib.performance.loadtest.profiles.Accepter</string>
-
- <key>params</key>
- <dict>
- <key>enabled</key>
- <false/>
-
- <!-- Define how long to wait after seeing a new invitation before
- accepting it.
-
- For LogNormal "mode" is the peak, "median" is the 50% cummulative value
- (i.e., half of the user have accepted by that time).
- -->
- <key>acceptDelayDistribution</key>
- <dict>
- <key>type</key>
- <string>contrib.performance.stats.LogNormalDistribution</string>
- <key>params</key>
- <dict>
- <!-- mode - peak-->
- <key>mode</key>
- <integer>300</integer>
- <!-- median - 50% done-->
- <key>median</key>
- <integer>1800</integer>
- </dict>
- </dict>
- </dict>
- </dict>
-
- <!-- A task-creating profile, which will periodically create
- new tasks at a random time on a random calendar. -->
- <dict>
- <key>class</key>
- <string>contrib.performance.loadtest.profiles.Tasker</string>
-
- <key>params</key>
- <dict>
- <key>enabled</key>
- <false/>
-
- <!-- Define the interval (in seconds) at which this profile will use
- its client to create a new task. -->
- <key>interval</key>
- <integer>300</integer>
-
- <!-- Define how due times (DUE) for the randomly generated tasks
- will be selected. This is an example of a "Distribution" parameter. The value
- for most "Distribution" parameters are interchangeable and extensible. -->
- <key>taskDueDistribution</key>
- <dict>
-
- <!-- This distribution is pretty specialized. It produces timestamps
- in the near future, limited to certain days of the week and certain hours
- of the day. -->
- <key>type</key>
- <string>contrib.performance.stats.WorkDistribution</string>
-
- <key>params</key>
- <dict>
- <!-- These are the days of the week the distribution will use. -->
- <key>daysOfWeek</key>
- <array>
- <string>mon</string>
- <string>tue</string>
- <string>wed</string>
- <string>thu</string>
- <string>fri</string>
- </array>
-
- <!-- The earliest hour of a day at which an event might be scheduled. -->
- <key>beginHour</key>
- <integer>8</integer>
-
- <!-- And the latest hour of a day (at which an event will be scheduled
- to begin!). -->
- <key>endHour</key>
- <integer>16</integer>
-
- <!-- The timezone in which the event is scheduled. (XXX Does this
- really work right?) -->
- <key>tzname</key>
- <string>America/Los_Angeles</string>
- </dict>
- </dict>
- </dict>
- </dict>
-
- </array>
-
- <!-- Determine the frequency at which this client configuration will
- appear in the clients which are created by the load tester. -->
- <key>weight</key>
- <integer>1</integer>
- </dict>
- </array>
- </dict>
-</plist>
Deleted: CalendarServer/trunk/contrib/performance/loadtest/standard-configs/events-only.plist
===================================================================
--- CalendarServer/trunk/contrib/performance/loadtest/standard-configs/events-only.plist 2016-06-22 21:47:05 UTC (rev 15690)
+++ CalendarServer/trunk/contrib/performance/loadtest/standard-configs/events-only.plist 2016-06-23 00:42:02 UTC (rev 15691)
@@ -1,543 +0,0 @@
-<?xml version="1.0" encoding="UTF-8"?>
-
-<!--
- Copyright (c) 2011-2016 Apple Inc. All rights reserved.
-
- Licensed under the Apache License, Version 2.0 (the "License");
- you may not use this file except in compliance with the License.
- You may obtain a copy of the License at
-
- http://www.apache.org/licenses/LICENSE-2.0
-
- Unless required by applicable law or agreed to in writing, software
- distributed under the License is distributed on an "AS IS" BASIS,
- WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- See the License for the specific language governing permissions and
- limitations under the License.
- -->
-
-<!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">
-<plist version="1.0">
- <dict>
- <!-- Define the kinds of software and user behavior the load simulation
- will simulate. -->
- <key>clients</key>
-
- <!-- Have as many different kinds of software and user behavior configurations
- as you want. Each is a dict -->
- <array>
-
- <dict>
-
- <!-- Here is a Lion iCal simulator. -->
- <key>software</key>
- <string>contrib.performance.loadtest.ical.OS_X_10_7</string>
-
- <!-- Arguments to use to initialize the client instance. -->
- <key>params</key>
- <dict>
- <!-- Name that appears in logs. -->
- <key>title</key>
- <string>10.7</string>
-
- <!-- Client can poll the calendar home at some interval. This is
- in seconds. -->
- <key>calendarHomePollInterval</key>
- <integer>300000</integer>
-
- <!-- If the server advertises xmpp push, OS X 10.6 can wait for notifications
- about calendar home changes instead of polling for them periodically. If
- this option is true, then look for the server advertisement for xmpp push
- and use it if possible. Still fall back to polling if there is no xmpp push
- advertised. -->
- <key>supportPush</key>
- <false />
- <key>supportAmpPush</key>
- <false />
- </dict>
-
- <!-- The profiles define certain types of user behavior on top of the
- client software being simulated. -->
- <key>profiles</key>
- <array>
-
- <!-- First an event-creating profile, which will periodically create
- new events at a random time on a random calendar. -->
- <dict>
- <key>class</key>
- <string>contrib.performance.loadtest.profiles.Eventer</string>
-
- <key>params</key>
- <dict>
- <key>enabled</key>
- <true/>
-
- <!-- Define the interval (in seconds) at which this profile will use
- its client to create a new event. -->
- <key>interval</key>
- <integer>20</integer>
-
- <!-- Define how start times (DTSTART) for the randomly generated events
- will be selected. This is an example of a "Distribution" parameter. The value
- for most "Distribution" parameters are interchangeable and extensible. -->
- <key>eventStartDistribution</key>
- <dict>
-
- <!-- This distribution is pretty specialized. It produces timestamps
- in the near future, limited to certain days of the week and certain hours
- of the day. -->
- <key>type</key>
- <string>contrib.performance.stats.WorkDistribution</string>
-
- <key>params</key>
- <dict>
- <!-- These are the days of the week the distribution will use. -->
- <key>daysOfWeek</key>
- <array>
- <string>mon</string>
- <string>tue</string>
- <string>wed</string>
- <string>thu</string>
- <string>fri</string>
- </array>
-
- <!-- The earliest hour of a day at which an event might be scheduled. -->
- <key>beginHour</key>
- <integer>8</integer>
-
- <!-- And the latest hour of a day (at which an event will be scheduled
- to begin!). -->
- <key>endHour</key>
- <integer>16</integer>
-
- <!-- The timezone in which the event is scheduled. (XXX Does this
- really work right?) -->
- <key>tzname</key>
- <string>America/Los_Angeles</string>
- </dict>
- </dict>
-
- <!-- Define how recurrences are created. -->
- <key>recurrenceDistribution</key>
- <dict>
-
- <!-- This distribution is pretty specialized. We have a fixed set of
- RRULEs defined for this distribution and pick each based on a
- weight. -->
- <key>type</key>
- <string>contrib.performance.stats.RecurrenceDistribution</string>
-
- <key>params</key>
- <dict>
- <!-- False to disable RRULEs -->
- <key>allowRecurrence</key>
- <false/>
-
- <!-- These are the weights for the specific set of RRULEs. -->
- <key>weights</key>
- <dict>
- <!-- Half of all events will be non-recurring -->
- <key>none</key>
- <integer>50</integer>
-
- <!-- Daily and weekly are pretty common -->
- <key>daily</key>
- <integer>10</integer>
- <key>weekly</key>
- <integer>20</integer>
-
- <!-- Monthly, yearly, daily & weekly limit not so common -->
- <key>monthly</key>
- <integer>2</integer>
- <key>yearly</key>
- <integer>1</integer>
- <key>dailylimit</key>
- <integer>2</integer>
- <key>weeklylimit</key>
- <integer>5</integer>
-
- <!-- Work days pretty common -->
- <key>workdays</key>
- <integer>10</integer>
- </dict>
- </dict>
- </dict>
- </dict>
- </dict>
-
- <!-- This profile will create a new event, and then periodically update the ACKNOWLEDGED property. -->
- <dict>
- <key>class</key>
- <string>contrib.performance.loadtest.profiles.EventUpdater</string>
-
- <key>params</key>
- <dict>
- <key>enabled</key>
- <false/>
-
- <!-- Define the interval (in seconds) at which this profile will use
- its client to create a new event. -->
- <key>interval</key>
- <integer>5</integer>
-
- <!-- Define how start times (DTSTART) for the randomly generated events
- will be selected. This is an example of a "Distribution" parameter. The value
- for most "Distribution" parameters are interchangeable and extensible. -->
- <key>eventStartDistribution</key>
- <dict>
-
- <!-- This distribution is pretty specialized. It produces timestamps
- in the near future, limited to certain days of the week and certain hours
- of the day. -->
- <key>type</key>
- <string>contrib.performance.stats.WorkDistribution</string>
-
- <key>params</key>
- <dict>
- <!-- These are the days of the week the distribution will use. -->
- <key>daysOfWeek</key>
- <array>
- <string>mon</string>
- <string>tue</string>
- <string>wed</string>
- <string>thu</string>
- <string>fri</string>
- </array>
-
- <!-- The earliest hour of a day at which an event might be scheduled. -->
- <key>beginHour</key>
- <integer>8</integer>
-
- <!-- And the latest hour of a day (at which an event will be scheduled
- to begin!). -->
- <key>endHour</key>
- <integer>16</integer>
-
- <!-- The timezone in which the event is scheduled. (XXX Does this
- really work right?) -->
- <key>tzname</key>
- <string>America/Los_Angeles</string>
- </dict>
- </dict>
-
- <!-- Define how recurrences are created. -->
- <key>recurrenceDistribution</key>
- <dict>
-
- <!-- This distribution is pretty specialized. We have a fixed set of
- RRULEs defined for this distribution and pick each based on a
- weight. -->
- <key>type</key>
- <string>contrib.performance.stats.RecurrenceDistribution</string>
-
- <key>params</key>
- <dict>
- <!-- False to disable RRULEs -->
- <key>allowRecurrence</key>
- <true/>
-
- <!-- These are the weights for the specific set of RRULEs. -->
- <key>weights</key>
- <dict>
- <!-- Half of all events will be non-recurring -->
- <key>none</key>
- <integer>50</integer>
-
- <!-- Daily and weekly are pretty common -->
- <key>daily</key>
- <integer>25</integer>
- <key>weekly</key>
- <integer>25</integer>
-
- <!-- Monthly, yearly, daily & weekly limit not so common -->
- <key>monthly</key>
- <integer>0</integer>
- <key>yearly</key>
- <integer>0</integer>
- <key>dailylimit</key>
- <integer>0</integer>
- <key>weeklylimit</key>
- <integer>0</integer>
-
- <!-- Work days pretty common -->
- <key>workdays</key>
- <integer>0</integer>
- </dict>
- </dict>
- </dict>
- </dict>
- </dict>
-
- <!-- This profile invites some number of new attendees to new events. -->
- <dict>
- <key>class</key>
- <string>contrib.performance.loadtest.profiles.RealisticInviter</string>
-
- <key>params</key>
- <dict>
- <key>enabled</key>
- <false/>
-
- <!-- Define the frequency at which new invitations will be sent out. -->
- <key>sendInvitationDistribution</key>
- <dict>
- <key>type</key>
- <string>contrib.performance.stats.NormalDistribution</string>
- <key>params</key>
- <dict>
- <!-- mu gives the mean of the normal distribution (in seconds). -->
- <key>mu</key>
- <integer>10</integer>
-
- <!-- and sigma gives its standard deviation. -->
- <key>sigma</key>
- <integer>5</integer>
- </dict>
- </dict>
-
- <!-- Define the distribution of who will be invited to an event.
-
- When inviteeClumping is turned on each invitee is based on a sample of
- users "close to" the organizer based on account index. If the clumping
- is too "tight" for the requested number of attendees, then invites for
- those larger numbers will simply fail (the sim will report that situation).
-
- When inviteeClumping is off invitees will be sampled across an entire
- range of account indexes. In this case the distribution ought to be a
- UniformIntegerDistribution with min=0 and max set to the number of accounts.
- -->
- <key>inviteeDistribution</key>
- <dict>
- <key>type</key>
- <string>contrib.performance.stats.UniformIntegerDistribution</string>
- <key>params</key>
- <dict>
- <!-- The minimum value (inclusive) of the uniform distribution. -->
- <key>min</key>
- <integer>0</integer>
- <!-- The maximum value (exclusive) of the uniform distribution. -->
- <key>max</key>
- <integer>99</integer>
- </dict>
- </dict>
-
- <key>inviteeClumping</key>
- <true/>
-
- <!-- Define the distribution of how many attendees will be invited to an event.
-
- LogNormal is the best fit to observed data.
-
-
- For LogNormal "mode" is the peak, "mean" is the mean value. For invites,
- mode should typically be 1, and mean whatever matches the user behavior.
- Our typical mean is 6.
- -->
- <key>inviteeCountDistribution</key>
- <dict>
- <key>type</key>
- <string>contrib.performance.stats.LogNormalDistribution</string>
- <key>params</key>
- <dict>
- <!-- mode - peak-->
- <key>mode</key>
- <integer>1</integer>
- <!-- mean - average-->
- <key>median</key>
- <integer>6</integer>
- <!-- maximum -->
- <key>maximum</key>
- <real>100</real>
- </dict>
- </dict>
-
- <!-- Define how start times (DTSTART) for the randomly generated events
- will be selected. This is an example of a "Distribution" parameter. The value
- for most "Distribution" parameters are interchangeable and extensible. -->
- <key>eventStartDistribution</key>
- <dict>
-
- <!-- This distribution is pretty specialized. It produces timestamps
- in the near future, limited to certain days of the week and certain hours
- of the day. -->
- <key>type</key>
- <string>contrib.performance.stats.WorkDistribution</string>
-
- <key>params</key>
- <dict>
- <!-- These are the days of the week the distribution will use. -->
- <key>daysOfWeek</key>
- <array>
- <string>mon</string>
- <string>tue</string>
- <string>wed</string>
- <string>thu</string>
- <string>fri</string>
- </array>
-
- <!-- The earliest hour of a day at which an event might be scheduled. -->
- <key>beginHour</key>
- <integer>8</integer>
-
- <!-- And the latest hour of a day (at which an event will be scheduled
- to begin!). -->
- <key>endHour</key>
- <integer>16</integer>
-
- <!-- The timezone in which the event is scheduled. (XXX Does this
- really work right?) -->
- <key>tzname</key>
- <string>America/Los_Angeles</string>
- </dict>
- </dict>
-
- <!-- Define how recurrences are created. -->
- <key>recurrenceDistribution</key>
- <dict>
-
- <!-- This distribution is pretty specialized. We have a fixed set of
- RRULEs defined for this distribution and pick each based on a
- weight. -->
- <key>type</key>
- <string>contrib.performance.stats.RecurrenceDistribution</string>
-
- <key>params</key>
- <dict>
- <!-- False to disable RRULEs -->
- <key>allowRecurrence</key>
- <true/>
-
- <!-- These are the weights for the specific set of RRULEs. -->
- <key>weights</key>
- <dict>
- <!-- Half of all events will be non-recurring -->
- <key>none</key>
- <integer>50</integer>
-
- <!-- Daily and weekly are pretty common -->
- <key>daily</key>
- <integer>10</integer>
- <key>weekly</key>
- <integer>20</integer>
-
- <!-- Monthly, yearly, daily & weekly limit not so common -->
- <key>monthly</key>
- <integer>2</integer>
- <key>yearly</key>
- <integer>1</integer>
- <key>dailylimit</key>
- <integer>2</integer>
- <key>weeklylimit</key>
- <integer>5</integer>
-
- <!-- Work days pretty common -->
- <key>workdays</key>
- <integer>10</integer>
- </dict>
- </dict>
- </dict>
- </dict>
- </dict>
-
- <!-- This profile accepts invitations to events, handles cancels, and
- handles replies received. -->
- <dict>
- <key>class</key>
- <string>contrib.performance.loadtest.profiles.Accepter</string>
-
- <key>params</key>
- <dict>
- <key>enabled</key>
- <false/>
-
- <!-- Define how long to wait after seeing a new invitation before
- accepting it.
-
- For LogNormal "mode" is the peak, "median" is the 50% cummulative value
- (i.e., half of the user have accepted by that time).
- -->
- <key>acceptDelayDistribution</key>
- <dict>
- <key>type</key>
- <string>contrib.performance.stats.LogNormalDistribution</string>
- <key>params</key>
- <dict>
- <!-- mode - peak-->
- <key>mode</key>
- <integer>300</integer>
- <!-- median - 50% done-->
- <key>median</key>
- <integer>1800</integer>
- </dict>
- </dict>
- </dict>
- </dict>
-
- <!-- A task-creating profile, which will periodically create
- new tasks at a random time on a random calendar. -->
- <dict>
- <key>class</key>
- <string>contrib.performance.loadtest.profiles.Tasker</string>
-
- <key>params</key>
- <dict>
- <key>enabled</key>
- <false/>
-
- <!-- Define the interval (in seconds) at which this profile will use
- its client to create a new task. -->
- <key>interval</key>
- <integer>300</integer>
-
- <!-- Define how due times (DUE) for the randomly generated tasks
- will be selected. This is an example of a "Distribution" parameter. The value
- for most "Distribution" parameters are interchangeable and extensible. -->
- <key>taskDueDistribution</key>
- <dict>
-
- <!-- This distribution is pretty specialized. It produces timestamps
- in the near future, limited to certain days of the week and certain hours
- of the day. -->
- <key>type</key>
- <string>contrib.performance.stats.WorkDistribution</string>
-
- <key>params</key>
- <dict>
- <!-- These are the days of the week the distribution will use. -->
- <key>daysOfWeek</key>
- <array>
- <string>mon</string>
- <string>tue</string>
- <string>wed</string>
- <string>thu</string>
- <string>fri</string>
- </array>
-
- <!-- The earliest hour of a day at which an event might be scheduled. -->
- <key>beginHour</key>
- <integer>8</integer>
-
- <!-- And the latest hour of a day (at which an event will be scheduled
- to begin!). -->
- <key>endHour</key>
- <integer>16</integer>
-
- <!-- The timezone in which the event is scheduled. (XXX Does this
- really work right?) -->
- <key>tzname</key>
- <string>America/Los_Angeles</string>
- </dict>
- </dict>
- </dict>
- </dict>
-
- </array>
-
- <!-- Determine the frequency at which this client configuration will
- appear in the clients which are created by the load tester. -->
- <key>weight</key>
- <integer>1</integer>
- </dict>
- </array>
- </dict>
-</plist>
Deleted: CalendarServer/trunk/contrib/performance/loadtest/standard-configs/invites-accepts.plist
===================================================================
--- CalendarServer/trunk/contrib/performance/loadtest/standard-configs/invites-accepts.plist 2016-06-22 21:47:05 UTC (rev 15690)
+++ CalendarServer/trunk/contrib/performance/loadtest/standard-configs/invites-accepts.plist 2016-06-23 00:42:02 UTC (rev 15691)
@@ -1,522 +0,0 @@
-<?xml version="1.0" encoding="UTF-8"?>
-
-<!--
- Copyright (c) 2011-2016 Apple Inc. All rights reserved.
-
- Licensed under the Apache License, Version 2.0 (the "License");
- you may not use this file except in compliance with the License.
- You may obtain a copy of the License at
-
- http://www.apache.org/licenses/LICENSE-2.0
-
- Unless required by applicable law or agreed to in writing, software
- distributed under the License is distributed on an "AS IS" BASIS,
- WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- See the License for the specific language governing permissions and
- limitations under the License.
- -->
-
-<!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">
-<plist version="1.0">
- <dict>
- <!-- Define the kinds of software and user behavior the load simulation
- will simulate. -->
- <key>clients</key>
-
- <!-- Have as many different kinds of software and user behavior configurations
- as you want. Each is a dict -->
- <array>
-
- <dict>
-
- <!-- Here is a Lion iCal simulator. -->
- <key>software</key>
- <string>contrib.performance.loadtest.ical.OS_X_10_7</string>
-
- <!-- Arguments to use to initialize the client instance. -->
- <key>params</key>
- <dict>
- <!-- Name that appears in logs. -->
- <key>title</key>
- <string>10.7</string>
-
- <!-- Client can poll the calendar home at some interval. This is
- in seconds. -->
- <key>calendarHomePollInterval</key>
- <integer>300000</integer>
-
- <!-- If the server advertises xmpp push, OS X 10.6 can wait for notifications
- about calendar home changes instead of polling for them periodically. If
- this option is true, then look for the server advertisement for xmpp push
- and use it if possible. Still fall back to polling if there is no xmpp push
- advertised. -->
- <key>supportPush</key>
- <false />
- <key>supportAmpPush</key>
- <true />
- </dict>
-
- <!-- The profiles define certain types of user behavior on top of the
- client software being simulated. -->
- <key>profiles</key>
- <array>
-
- <!-- First an event-creating profile, which will periodically create
- new events at a random time on a random calendar. -->
- <dict>
- <key>class</key>
- <string>contrib.performance.loadtest.profiles.Eventer</string>
-
- <key>params</key>
- <dict>
- <key>enabled</key>
- <false/>
-
- <!-- Define the interval (in seconds) at which this profile will use
- its client to create a new event. -->
- <key>interval</key>
- <integer>20</integer>
-
- <!-- Define how start times (DTSTART) for the randomly generated events
- will be selected. This is an example of a "Distribution" parameter. The value
- for most "Distribution" parameters are interchangeable and extensible. -->
- <key>eventStartDistribution</key>
- <dict>
-
- <!-- This distribution is pretty specialized. It produces timestamps
- in the near future, limited to certain days of the week and certain hours
- of the day. -->
- <key>type</key>
- <string>contrib.performance.stats.WorkDistribution</string>
-
- <key>params</key>
- <dict>
- <!-- These are the days of the week the distribution will use. -->
- <key>daysOfWeek</key>
- <array>
- <string>mon</string>
- <string>tue</string>
- <string>wed</string>
- <string>thu</string>
- <string>fri</string>
- </array>
-
- <!-- The earliest hour of a day at which an event might be scheduled. -->
- <key>beginHour</key>
- <integer>8</integer>
-
- <!-- And the latest hour of a day (at which an event will be scheduled
- to begin!). -->
- <key>endHour</key>
- <integer>16</integer>
-
- <!-- The timezone in which the event is scheduled. (XXX Does this
- really work right?) -->
- <key>tzname</key>
- <string>America/Los_Angeles</string>
- </dict>
- </dict>
-
- <!-- Define how recurrences are created. -->
- <key>recurrenceDistribution</key>
- <dict>
-
- <!-- This distribution is pretty specialized. We have a fixed set of
- RRULEs defined for this distribution and pick each based on a
- weight. -->
- <key>type</key>
- <string>contrib.performance.stats.RecurrenceDistribution</string>
-
- <key>params</key>
- <dict>
- <!-- False to disable RRULEs -->
- <key>allowRecurrence</key>
- <false/>
-
- <!-- These are the weights for the specific set of RRULEs. -->
- <key>weights</key>
- <dict>
- <!-- Half of all events will be non-recurring -->
- <key>none</key>
- <integer>50</integer>
-
- <!-- Daily and weekly are pretty common -->
- <key>daily</key>
- <integer>10</integer>
- <key>weekly</key>
- <integer>20</integer>
-
- <!-- Monthly, yearly, daily & weekly limit not so common -->
- <key>monthly</key>
- <integer>2</integer>
- <key>yearly</key>
- <integer>1</integer>
- <key>dailylimit</key>
- <integer>2</integer>
- <key>weeklylimit</key>
- <integer>5</integer>
-
- <!-- Work days pretty common -->
- <key>workdays</key>
- <integer>10</integer>
- </dict>
- </dict>
- </dict>
- </dict>
- </dict>
-
- <!-- This profile will create a new event, and then periodically update the ACKNOWLEDGED property. -->
- <dict>
- <key>class</key>
- <string>contrib.performance.loadtest.profiles.EventUpdater</string>
-
- <key>params</key>
- <dict>
- <key>enabled</key>
- <false/>
-
- <!-- Define the interval (in seconds) at which this profile will use
- its client to create a new event. -->
- <key>interval</key>
- <integer>5</integer>
-
- <!-- Define how start times (DTSTART) for the randomly generated events
- will be selected. This is an example of a "Distribution" parameter. The value
- for most "Distribution" parameters are interchangeable and extensible. -->
- <key>eventStartDistribution</key>
- <dict>
-
- <!-- This distribution is pretty specialized. It produces timestamps
- in the near future, limited to certain days of the week and certain hours
- of the day. -->
- <key>type</key>
- <string>contrib.performance.stats.WorkDistribution</string>
-
- <key>params</key>
- <dict>
- <!-- These are the days of the week the distribution will use. -->
- <key>daysOfWeek</key>
- <array>
- <string>mon</string>
- <string>tue</string>
- <string>wed</string>
- <string>thu</string>
- <string>fri</string>
- </array>
-
- <!-- The earliest hour of a day at which an event might be scheduled. -->
- <key>beginHour</key>
- <integer>8</integer>
-
- <!-- And the latest hour of a day (at which an event will be scheduled
- to begin!). -->
- <key>endHour</key>
- <integer>16</integer>
-
- <!-- The timezone in which the event is scheduled. (XXX Does this
- really work right?) -->
- <key>tzname</key>
- <string>America/Los_Angeles</string>
- </dict>
- </dict>
-
- <!-- Define how recurrences are created. -->
- <key>recurrenceDistribution</key>
- <dict>
-
- <!-- This distribution is pretty specialized. We have a fixed set of
- RRULEs defined for this distribution and pick each based on a
- weight. -->
- <key>type</key>
- <string>contrib.performance.stats.RecurrenceDistribution</string>
-
- <key>params</key>
- <dict>
- <!-- False to disable RRULEs -->
- <key>allowRecurrence</key>
- <true/>
-
- <!-- These are the weights for the specific set of RRULEs. -->
- <key>weights</key>
- <dict>
- <!-- Half of all events will be non-recurring -->
- <key>none</key>
- <integer>50</integer>
-
- <!-- Daily and weekly are pretty common -->
- <key>daily</key>
- <integer>25</integer>
- <key>weekly</key>
- <integer>25</integer>
-
- <!-- Monthly, yearly, daily & weekly limit not so common -->
- <key>monthly</key>
- <integer>0</integer>
- <key>yearly</key>
- <integer>0</integer>
- <key>dailylimit</key>
- <integer>0</integer>
- <key>weeklylimit</key>
- <integer>0</integer>
-
- <!-- Work days pretty common -->
- <key>workdays</key>
- <integer>0</integer>
- </dict>
- </dict>
- </dict>
- </dict>
- </dict>
-
- <!-- This profile invites some number of new attendees to new events. -->
- <dict>
- <key>class</key>
- <string>contrib.performance.loadtest.profiles.RealisticInviter</string>
-
- <key>params</key>
- <dict>
- <key>enabled</key>
- <true/>
-
- <!-- Define the frequency at which new invitations will be sent out. -->
- <key>sendInvitationDistribution</key>
- <dict>
- <key>type</key>
- <string>contrib.performance.stats.FixedDistribution</string>
- <key>params</key>
- <dict>
- <!-- interval (in seconds). -->
- <key>value</key>
- <integer>150</integer>
- </dict>
- </dict>
-
- <!-- Define the distribution of who will be invited to an event.
-
- When inviteeClumping is turned on each invitee is based on a sample of
- users "close to" the organizer based on account index. If the clumping
- is too "tight" for the requested number of attendees, then invites for
- those larger numbers will simply fail (the sim will report that situation).
-
- When inviteeClumping is off invitees will be sampled across an entire
- range of account indexes. In this case the distribution ought to be a
- UniformIntegerDistribution with min=0 and max set to the number of accounts.
- -->
- <key>inviteeDistribution</key>
- <dict>
- <key>type</key>
- <string>contrib.performance.stats.UniformIntegerDistribution</string>
- <key>params</key>
- <dict>
- <!-- The minimum value (inclusive) of the uniform distribution. -->
- <key>min</key>
- <integer>0</integer>
- <!-- The maximum value (exclusive) of the uniform distribution. -->
- <key>max</key>
- <integer>99</integer>
- </dict>
- </dict>
-
- <key>inviteeClumping</key>
- <true/>
-
- <!-- Define the distribution of how many attendees will be invited to an event.
-
- LogNormal is the best fit to observed data.
-
-
- For LogNormal "mode" is the peak, "mean" is the mean value. For invites,
- mode should typically be 1, and mean whatever matches the user behavior.
- Our typical mean is 6.
- -->
- <key>inviteeCountDistribution</key>
- <dict>
- <key>type</key>
- <string>contrib.performance.stats.FixedDistribution</string>
- <key>params</key>
- <dict>
- <!-- Number of attendees. -->
- <key>value</key>
- <integer>5</integer>
- </dict>
- </dict>
-
- <!-- Define how start times (DTSTART) for the randomly generated events
- will be selected. This is an example of a "Distribution" parameter. The value
- for most "Distribution" parameters are interchangeable and extensible. -->
- <key>eventStartDistribution</key>
- <dict>
-
- <!-- This distribution is pretty specialized. It produces timestamps
- in the near future, limited to certain days of the week and certain hours
- of the day. -->
- <key>type</key>
- <string>contrib.performance.stats.WorkDistribution</string>
-
- <key>params</key>
- <dict>
- <!-- These are the days of the week the distribution will use. -->
- <key>daysOfWeek</key>
- <array>
- <string>mon</string>
- <string>tue</string>
- <string>wed</string>
- <string>thu</string>
- <string>fri</string>
- </array>
-
- <!-- The earliest hour of a day at which an event might be scheduled. -->
- <key>beginHour</key>
- <integer>8</integer>
-
- <!-- And the latest hour of a day (at which an event will be scheduled
- to begin!). -->
- <key>endHour</key>
- <integer>16</integer>
-
- <!-- The timezone in which the event is scheduled. (XXX Does this
- really work right?) -->
- <key>tzname</key>
- <string>America/Los_Angeles</string>
- </dict>
- </dict>
-
- <!-- Define how recurrences are created. -->
- <key>recurrenceDistribution</key>
- <dict>
-
- <!-- This distribution is pretty specialized. We have a fixed set of
- RRULEs defined for this distribution and pick each based on a
- weight. -->
- <key>type</key>
- <string>contrib.performance.stats.RecurrenceDistribution</string>
-
- <key>params</key>
- <dict>
- <!-- False to disable RRULEs -->
- <key>allowRecurrence</key>
- <false/>
-
- <!-- These are the weights for the specific set of RRULEs. -->
- <key>weights</key>
- <dict>
- <!-- Daily and weekly are pretty common -->
- <key>daily</key>
- <integer>100</integer>
- </dict>
- </dict>
- </dict>
- </dict>
- </dict>
-
- <!-- This profile accepts invitations to events, handles cancels, and
- handles replies received. -->
- <dict>
- <key>class</key>
- <string>contrib.performance.loadtest.profiles.Accepter</string>
-
- <key>params</key>
- <dict>
- <key>enabled</key>
- <true/>
-
- <!-- Define how long to wait after seeing a new invitation before
- accepting it.
-
- For LogNormal "mode" is the peak, "median" is the 50% cummulative value
- (i.e., half of the user have accepted by that time).
- -->
- <key>acceptDelayDistribution</key>
- <dict>
- <key>type</key>
- <string>contrib.performance.stats.UniformDiscreteDistribution</string>
- <key>params</key>
- <dict>
- <!-- Set of values to use - will be chosen in random order. -->
- <key>values</key>
- <array>
- <integer>0</integer>
- <integer>5</integer>
- <integer>10</integer>
- <integer>15</integer>
- <integer>20</integer>
- <integer>25</integer>
- <integer>30</integer>
- </array>
- </dict>
- </dict>
- </dict>
- </dict>
-
- <!-- A task-creating profile, which will periodically create
- new tasks at a random time on a random calendar. -->
- <dict>
- <key>class</key>
- <string>contrib.performance.loadtest.profiles.Tasker</string>
-
- <key>params</key>
- <dict>
- <key>enabled</key>
- <false/>
-
- <!-- Define the interval (in seconds) at which this profile will use
- its client to create a new task. -->
- <key>interval</key>
- <integer>300</integer>
-
- <!-- Define how due times (DUE) for the randomly generated tasks
- will be selected. This is an example of a "Distribution" parameter. The value
- for most "Distribution" parameters are interchangeable and extensible. -->
- <key>taskDueDistribution</key>
- <dict>
-
- <!-- This distribution is pretty specialized. It produces timestamps
- in the near future, limited to certain days of the week and certain hours
- of the day. -->
- <key>type</key>
- <string>contrib.performance.stats.WorkDistribution</string>
-
- <key>params</key>
- <dict>
- <!-- These are the days of the week the distribution will use. -->
- <key>daysOfWeek</key>
- <array>
- <string>mon</string>
- <string>tue</string>
- <string>wed</string>
- <string>thu</string>
- <string>fri</string>
- </array>
-
- <!-- The earliest hour of a day at which an event might be scheduled. -->
- <key>beginHour</key>
- <integer>8</integer>
-
- <!-- And the latest hour of a day (at which an event will be scheduled
- to begin!). -->
- <key>endHour</key>
- <integer>16</integer>
-
- <!-- The timezone in which the event is scheduled. (XXX Does this
- really work right?) -->
- <key>tzname</key>
- <string>America/Los_Angeles</string>
- </dict>
- </dict>
- </dict>
- </dict>
-
- </array>
-
- <!-- Determine the frequency at which this client configuration will
- appear in the clients which are created by the load tester. -->
- <key>weight</key>
- <integer>1</integer>
- </dict>
- </array>
-
- <!-- Determine the interval between client creation. -->
- <key>arrivalInterval</key>
- <integer>5</integer>
- </dict>
-</plist>
Deleted: CalendarServer/trunk/contrib/performance/loadtest/standard-configs/invites-only-recurring.plist
===================================================================
--- CalendarServer/trunk/contrib/performance/loadtest/standard-configs/invites-only-recurring.plist 2016-06-22 21:47:05 UTC (rev 15690)
+++ CalendarServer/trunk/contrib/performance/loadtest/standard-configs/invites-only-recurring.plist 2016-06-23 00:42:02 UTC (rev 15691)
@@ -1,517 +0,0 @@
-<?xml version="1.0" encoding="UTF-8"?>
-
-<!--
- Copyright (c) 2011-2016 Apple Inc. All rights reserved.
-
- Licensed under the Apache License, Version 2.0 (the "License");
- you may not use this file except in compliance with the License.
- You may obtain a copy of the License at
-
- http://www.apache.org/licenses/LICENSE-2.0
-
- Unless required by applicable law or agreed to in writing, software
- distributed under the License is distributed on an "AS IS" BASIS,
- WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- See the License for the specific language governing permissions and
- limitations under the License.
- -->
-
-<!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">
-<plist version="1.0">
- <dict>
- <!-- Define the kinds of software and user behavior the load simulation
- will simulate. -->
- <key>clients</key>
-
- <!-- Have as many different kinds of software and user behavior configurations
- as you want. Each is a dict -->
- <array>
-
- <dict>
-
- <!-- Here is a Lion iCal simulator. -->
- <key>software</key>
- <string>contrib.performance.loadtest.ical.OS_X_10_7</string>
-
- <!-- Arguments to use to initialize the client instance. -->
- <key>params</key>
- <dict>
- <!-- Name that appears in logs. -->
- <key>title</key>
- <string>10.7</string>
-
- <!-- Client can poll the calendar home at some interval. This is
- in seconds. -->
- <key>calendarHomePollInterval</key>
- <integer>300000</integer>
-
- <!-- If the server advertises xmpp push, OS X 10.6 can wait for notifications
- about calendar home changes instead of polling for them periodically. If
- this option is true, then look for the server advertisement for xmpp push
- and use it if possible. Still fall back to polling if there is no xmpp push
- advertised. -->
- <key>supportPush</key>
- <false />
- <key>supportAmpPush</key>
- <false />
- </dict>
-
- <!-- The profiles define certain types of user behavior on top of the
- client software being simulated. -->
- <key>profiles</key>
- <array>
-
- <!-- First an event-creating profile, which will periodically create
- new events at a random time on a random calendar. -->
- <dict>
- <key>class</key>
- <string>contrib.performance.loadtest.profiles.Eventer</string>
-
- <key>params</key>
- <dict>
- <key>enabled</key>
- <false/>
-
- <!-- Define the interval (in seconds) at which this profile will use
- its client to create a new event. -->
- <key>interval</key>
- <integer>20</integer>
-
- <!-- Define how start times (DTSTART) for the randomly generated events
- will be selected. This is an example of a "Distribution" parameter. The value
- for most "Distribution" parameters are interchangeable and extensible. -->
- <key>eventStartDistribution</key>
- <dict>
-
- <!-- This distribution is pretty specialized. It produces timestamps
- in the near future, limited to certain days of the week and certain hours
- of the day. -->
- <key>type</key>
- <string>contrib.performance.stats.WorkDistribution</string>
-
- <key>params</key>
- <dict>
- <!-- These are the days of the week the distribution will use. -->
- <key>daysOfWeek</key>
- <array>
- <string>mon</string>
- <string>tue</string>
- <string>wed</string>
- <string>thu</string>
- <string>fri</string>
- </array>
-
- <!-- The earliest hour of a day at which an event might be scheduled. -->
- <key>beginHour</key>
- <integer>8</integer>
-
- <!-- And the latest hour of a day (at which an event will be scheduled
- to begin!). -->
- <key>endHour</key>
- <integer>16</integer>
-
- <!-- The timezone in which the event is scheduled. (XXX Does this
- really work right?) -->
- <key>tzname</key>
- <string>America/Los_Angeles</string>
- </dict>
- </dict>
-
- <!-- Define how recurrences are created. -->
- <key>recurrenceDistribution</key>
- <dict>
-
- <!-- This distribution is pretty specialized. We have a fixed set of
- RRULEs defined for this distribution and pick each based on a
- weight. -->
- <key>type</key>
- <string>contrib.performance.stats.RecurrenceDistribution</string>
-
- <key>params</key>
- <dict>
- <!-- False to disable RRULEs -->
- <key>allowRecurrence</key>
- <false/>
-
- <!-- These are the weights for the specific set of RRULEs. -->
- <key>weights</key>
- <dict>
- <!-- Half of all events will be non-recurring -->
- <key>none</key>
- <integer>50</integer>
-
- <!-- Daily and weekly are pretty common -->
- <key>daily</key>
- <integer>10</integer>
- <key>weekly</key>
- <integer>20</integer>
-
- <!-- Monthly, yearly, daily & weekly limit not so common -->
- <key>monthly</key>
- <integer>2</integer>
- <key>yearly</key>
- <integer>1</integer>
- <key>dailylimit</key>
- <integer>2</integer>
- <key>weeklylimit</key>
- <integer>5</integer>
-
- <!-- Work days pretty common -->
- <key>workdays</key>
- <integer>10</integer>
- </dict>
- </dict>
- </dict>
- </dict>
- </dict>
-
- <!-- This profile will create a new event, and then periodically update the ACKNOWLEDGED property. -->
- <dict>
- <key>class</key>
- <string>contrib.performance.loadtest.profiles.EventUpdater</string>
-
- <key>params</key>
- <dict>
- <key>enabled</key>
- <false/>
-
- <!-- Define the interval (in seconds) at which this profile will use
- its client to create a new event. -->
- <key>interval</key>
- <integer>5</integer>
-
- <!-- Define how start times (DTSTART) for the randomly generated events
- will be selected. This is an example of a "Distribution" parameter. The value
- for most "Distribution" parameters are interchangeable and extensible. -->
- <key>eventStartDistribution</key>
- <dict>
-
- <!-- This distribution is pretty specialized. It produces timestamps
- in the near future, limited to certain days of the week and certain hours
- of the day. -->
- <key>type</key>
- <string>contrib.performance.stats.WorkDistribution</string>
-
- <key>params</key>
- <dict>
- <!-- These are the days of the week the distribution will use. -->
- <key>daysOfWeek</key>
- <array>
- <string>mon</string>
- <string>tue</string>
- <string>wed</string>
- <string>thu</string>
- <string>fri</string>
- </array>
-
- <!-- The earliest hour of a day at which an event might be scheduled. -->
- <key>beginHour</key>
- <integer>8</integer>
-
- <!-- And the latest hour of a day (at which an event will be scheduled
- to begin!). -->
- <key>endHour</key>
- <integer>16</integer>
-
- <!-- The timezone in which the event is scheduled. (XXX Does this
- really work right?) -->
- <key>tzname</key>
- <string>America/Los_Angeles</string>
- </dict>
- </dict>
-
- <!-- Define how recurrences are created. -->
- <key>recurrenceDistribution</key>
- <dict>
-
- <!-- This distribution is pretty specialized. We have a fixed set of
- RRULEs defined for this distribution and pick each based on a
- weight. -->
- <key>type</key>
- <string>contrib.performance.stats.RecurrenceDistribution</string>
-
- <key>params</key>
- <dict>
- <!-- False to disable RRULEs -->
- <key>allowRecurrence</key>
- <true/>
-
- <!-- These are the weights for the specific set of RRULEs. -->
- <key>weights</key>
- <dict>
- <!-- Half of all events will be non-recurring -->
- <key>none</key>
- <integer>50</integer>
-
- <!-- Daily and weekly are pretty common -->
- <key>daily</key>
- <integer>25</integer>
- <key>weekly</key>
- <integer>25</integer>
-
- <!-- Monthly, yearly, daily & weekly limit not so common -->
- <key>monthly</key>
- <integer>0</integer>
- <key>yearly</key>
- <integer>0</integer>
- <key>dailylimit</key>
- <integer>0</integer>
- <key>weeklylimit</key>
- <integer>0</integer>
-
- <!-- Work days pretty common -->
- <key>workdays</key>
- <integer>0</integer>
- </dict>
- </dict>
- </dict>
- </dict>
- </dict>
-
- <!-- This profile invites some number of new attendees to new events. -->
- <dict>
- <key>class</key>
- <string>contrib.performance.loadtest.profiles.RealisticInviter</string>
-
- <key>params</key>
- <dict>
- <key>enabled</key>
- <true/>
-
- <!-- Define the frequency at which new invitations will be sent out. -->
- <key>sendInvitationDistribution</key>
- <dict>
- <key>type</key>
- <string>contrib.performance.stats.FixedDistribution</string>
- <key>params</key>
- <dict>
- <!-- interval (in seconds). -->
- <key>value</key>
- <integer>120</integer>
- </dict>
- </dict>
-
- <!-- Define the distribution of who will be invited to an event.
-
- When inviteeClumping is turned on each invitee is based on a sample of
- users "close to" the organizer based on account index. If the clumping
- is too "tight" for the requested number of attendees, then invites for
- those larger numbers will simply fail (the sim will report that situation).
-
- When inviteeClumping is off invitees will be sampled across an entire
- range of account indexes. In this case the distribution ought to be a
- UniformIntegerDistribution with min=0 and max set to the number of accounts.
- -->
- <key>inviteeDistribution</key>
- <dict>
- <key>type</key>
- <string>contrib.performance.stats.UniformIntegerDistribution</string>
- <key>params</key>
- <dict>
- <!-- The minimum value (inclusive) of the uniform distribution. -->
- <key>min</key>
- <integer>0</integer>
- <!-- The maximum value (exclusive) of the uniform distribution. -->
- <key>max</key>
- <integer>99</integer>
- </dict>
- </dict>
-
- <key>inviteeClumping</key>
- <true/>
-
- <!-- Define the distribution of how many attendees will be invited to an event.
-
- LogNormal is the best fit to observed data.
-
-
- For LogNormal "mode" is the peak, "mean" is the mean value. For invites,
- mode should typically be 1, and mean whatever matches the user behavior.
- Our typical mean is 6.
- -->
- <key>inviteeCountDistribution</key>
- <dict>
- <key>type</key>
- <string>contrib.performance.stats.FixedDistribution</string>
- <key>params</key>
- <dict>
- <!-- Number of attendees. -->
- <key>value</key>
- <integer>5</integer>
- </dict>
- </dict>
-
- <!-- Define how start times (DTSTART) for the randomly generated events
- will be selected. This is an example of a "Distribution" parameter. The value
- for most "Distribution" parameters are interchangeable and extensible. -->
- <key>eventStartDistribution</key>
- <dict>
-
- <!-- This distribution is pretty specialized. It produces timestamps
- in the near future, limited to certain days of the week and certain hours
- of the day. -->
- <key>type</key>
- <string>contrib.performance.stats.WorkDistribution</string>
-
- <key>params</key>
- <dict>
- <!-- These are the days of the week the distribution will use. -->
- <key>daysOfWeek</key>
- <array>
- <string>mon</string>
- <string>tue</string>
- <string>wed</string>
- <string>thu</string>
- <string>fri</string>
- </array>
-
- <!-- The earliest hour of a day at which an event might be scheduled. -->
- <key>beginHour</key>
- <integer>8</integer>
-
- <!-- And the latest hour of a day (at which an event will be scheduled
- to begin!). -->
- <key>endHour</key>
- <integer>16</integer>
-
- <!-- The timezone in which the event is scheduled. (XXX Does this
- really work right?) -->
- <key>tzname</key>
- <string>America/Los_Angeles</string>
- </dict>
- </dict>
-
- <!-- Define how recurrences are created. -->
- <key>recurrenceDistribution</key>
- <dict>
-
- <!-- This distribution is pretty specialized. We have a fixed set of
- RRULEs defined for this distribution and pick each based on a
- weight. -->
- <key>type</key>
- <string>contrib.performance.stats.RecurrenceDistribution</string>
-
- <key>params</key>
- <dict>
- <!-- False to disable RRULEs -->
- <key>allowRecurrence</key>
- <true/>
-
- <!-- These are the weights for the specific set of RRULEs. -->
- <key>weights</key>
- <dict>
- <!-- Daily and weekly are pretty common -->
- <key>daily</key>
- <integer>100</integer>
- </dict>
- </dict>
- </dict>
- </dict>
- </dict>
-
- <!-- This profile accepts invitations to events, handles cancels, and
- handles replies received. -->
- <dict>
- <key>class</key>
- <string>contrib.performance.loadtest.profiles.Accepter</string>
-
- <key>params</key>
- <dict>
- <key>enabled</key>
- <false/>
-
- <!-- Define how long to wait after seeing a new invitation before
- accepting it.
-
- For LogNormal "mode" is the peak, "median" is the 50% cummulative value
- (i.e., half of the user have accepted by that time).
- -->
- <key>acceptDelayDistribution</key>
- <dict>
- <key>type</key>
- <string>contrib.performance.stats.LogNormalDistribution</string>
- <key>params</key>
- <dict>
- <!-- mode - peak-->
- <key>mode</key>
- <integer>300</integer>
- <!-- median - 50% done-->
- <key>median</key>
- <integer>1800</integer>
- </dict>
- </dict>
- </dict>
- </dict>
-
- <!-- A task-creating profile, which will periodically create
- new tasks at a random time on a random calendar. -->
- <dict>
- <key>class</key>
- <string>contrib.performance.loadtest.profiles.Tasker</string>
-
- <key>params</key>
- <dict>
- <key>enabled</key>
- <false/>
-
- <!-- Define the interval (in seconds) at which this profile will use
- its client to create a new task. -->
- <key>interval</key>
- <integer>300</integer>
-
- <!-- Define how due times (DUE) for the randomly generated tasks
- will be selected. This is an example of a "Distribution" parameter. The value
- for most "Distribution" parameters are interchangeable and extensible. -->
- <key>taskDueDistribution</key>
- <dict>
-
- <!-- This distribution is pretty specialized. It produces timestamps
- in the near future, limited to certain days of the week and certain hours
- of the day. -->
- <key>type</key>
- <string>contrib.performance.stats.WorkDistribution</string>
-
- <key>params</key>
- <dict>
- <!-- These are the days of the week the distribution will use. -->
- <key>daysOfWeek</key>
- <array>
- <string>mon</string>
- <string>tue</string>
- <string>wed</string>
- <string>thu</string>
- <string>fri</string>
- </array>
-
- <!-- The earliest hour of a day at which an event might be scheduled. -->
- <key>beginHour</key>
- <integer>8</integer>
-
- <!-- And the latest hour of a day (at which an event will be scheduled
- to begin!). -->
- <key>endHour</key>
- <integer>16</integer>
-
- <!-- The timezone in which the event is scheduled. (XXX Does this
- really work right?) -->
- <key>tzname</key>
- <string>America/Los_Angeles</string>
- </dict>
- </dict>
- </dict>
- </dict>
-
- </array>
-
- <!-- Determine the frequency at which this client configuration will
- appear in the clients which are created by the load tester. -->
- <key>weight</key>
- <integer>1</integer>
- </dict>
- </array>
-
- <!-- Determine the interval between client creation. -->
- <key>arrivalInterval</key>
- <integer>4</integer>
- </dict>
-</plist>
Deleted: CalendarServer/trunk/contrib/performance/loadtest/standard-configs/invites-only.plist
===================================================================
--- CalendarServer/trunk/contrib/performance/loadtest/standard-configs/invites-only.plist 2016-06-22 21:47:05 UTC (rev 15690)
+++ CalendarServer/trunk/contrib/performance/loadtest/standard-configs/invites-only.plist 2016-06-23 00:42:02 UTC (rev 15691)
@@ -1,533 +0,0 @@
-<?xml version="1.0" encoding="UTF-8"?>
-
-<!--
- Copyright (c) 2011-2016 Apple Inc. All rights reserved.
-
- Licensed under the Apache License, Version 2.0 (the "License");
- you may not use this file except in compliance with the License.
- You may obtain a copy of the License at
-
- http://www.apache.org/licenses/LICENSE-2.0
-
- Unless required by applicable law or agreed to in writing, software
- distributed under the License is distributed on an "AS IS" BASIS,
- WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- See the License for the specific language governing permissions and
- limitations under the License.
- -->
-
-<!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">
-<plist version="1.0">
- <dict>
- <!-- Define the kinds of software and user behavior the load simulation
- will simulate. -->
- <key>clients</key>
-
- <!-- Have as many different kinds of software and user behavior configurations
- as you want. Each is a dict -->
- <array>
-
- <dict>
-
- <!-- Here is a Lion iCal simulator. -->
- <key>software</key>
- <string>contrib.performance.loadtest.ical.OS_X_10_7</string>
-
- <!-- Arguments to use to initialize the client instance. -->
- <key>params</key>
- <dict>
- <!-- Name that appears in logs. -->
- <key>title</key>
- <string>10.7</string>
-
- <!-- Client can poll the calendar home at some interval. This is
- in seconds. -->
- <key>calendarHomePollInterval</key>
- <integer>300000</integer>
-
- <!-- If the server advertises xmpp push, OS X 10.6 can wait for notifications
- about calendar home changes instead of polling for them periodically. If
- this option is true, then look for the server advertisement for xmpp push
- and use it if possible. Still fall back to polling if there is no xmpp push
- advertised. -->
- <key>supportPush</key>
- <false />
- <key>supportAmpPush</key>
- <false />
- </dict>
-
- <!-- The profiles define certain types of user behavior on top of the
- client software being simulated. -->
- <key>profiles</key>
- <array>
-
- <!-- First an event-creating profile, which will periodically create
- new events at a random time on a random calendar. -->
- <dict>
- <key>class</key>
- <string>contrib.performance.loadtest.profiles.Eventer</string>
-
- <key>params</key>
- <dict>
- <key>enabled</key>
- <false/>
-
- <!-- Define the interval (in seconds) at which this profile will use
- its client to create a new event. -->
- <key>interval</key>
- <integer>20</integer>
-
- <!-- Define how start times (DTSTART) for the randomly generated events
- will be selected. This is an example of a "Distribution" parameter. The value
- for most "Distribution" parameters are interchangeable and extensible. -->
- <key>eventStartDistribution</key>
- <dict>
-
- <!-- This distribution is pretty specialized. It produces timestamps
- in the near future, limited to certain days of the week and certain hours
- of the day. -->
- <key>type</key>
- <string>contrib.performance.stats.WorkDistribution</string>
-
- <key>params</key>
- <dict>
- <!-- These are the days of the week the distribution will use. -->
- <key>daysOfWeek</key>
- <array>
- <string>mon</string>
- <string>tue</string>
- <string>wed</string>
- <string>thu</string>
- <string>fri</string>
- </array>
-
- <!-- The earliest hour of a day at which an event might be scheduled. -->
- <key>beginHour</key>
- <integer>8</integer>
-
- <!-- And the latest hour of a day (at which an event will be scheduled
- to begin!). -->
- <key>endHour</key>
- <integer>16</integer>
-
- <!-- The timezone in which the event is scheduled. (XXX Does this
- really work right?) -->
- <key>tzname</key>
- <string>America/Los_Angeles</string>
- </dict>
- </dict>
-
- <!-- Define how recurrences are created. -->
- <key>recurrenceDistribution</key>
- <dict>
-
- <!-- This distribution is pretty specialized. We have a fixed set of
- RRULEs defined for this distribution and pick each based on a
- weight. -->
- <key>type</key>
- <string>contrib.performance.stats.RecurrenceDistribution</string>
-
- <key>params</key>
- <dict>
- <!-- False to disable RRULEs -->
- <key>allowRecurrence</key>
- <false/>
-
- <!-- These are the weights for the specific set of RRULEs. -->
- <key>weights</key>
- <dict>
- <!-- Half of all events will be non-recurring -->
- <key>none</key>
- <integer>50</integer>
-
- <!-- Daily and weekly are pretty common -->
- <key>daily</key>
- <integer>10</integer>
- <key>weekly</key>
- <integer>20</integer>
-
- <!-- Monthly, yearly, daily & weekly limit not so common -->
- <key>monthly</key>
- <integer>2</integer>
- <key>yearly</key>
- <integer>1</integer>
- <key>dailylimit</key>
- <integer>2</integer>
- <key>weeklylimit</key>
- <integer>5</integer>
-
- <!-- Work days pretty common -->
- <key>workdays</key>
- <integer>10</integer>
- </dict>
- </dict>
- </dict>
- </dict>
- </dict>
-
- <!-- This profile will create a new event, and then periodically update the ACKNOWLEDGED property. -->
- <dict>
- <key>class</key>
- <string>contrib.performance.loadtest.profiles.EventUpdater</string>
-
- <key>params</key>
- <dict>
- <key>enabled</key>
- <false/>
-
- <!-- Define the interval (in seconds) at which this profile will use
- its client to create a new event. -->
- <key>interval</key>
- <integer>5</integer>
-
- <!-- Define how start times (DTSTART) for the randomly generated events
- will be selected. This is an example of a "Distribution" parameter. The value
- for most "Distribution" parameters are interchangeable and extensible. -->
- <key>eventStartDistribution</key>
- <dict>
-
- <!-- This distribution is pretty specialized. It produces timestamps
- in the near future, limited to certain days of the week and certain hours
- of the day. -->
- <key>type</key>
- <string>contrib.performance.stats.WorkDistribution</string>
-
- <key>params</key>
- <dict>
- <!-- These are the days of the week the distribution will use. -->
- <key>daysOfWeek</key>
- <array>
- <string>mon</string>
- <string>tue</string>
- <string>wed</string>
- <string>thu</string>
- <string>fri</string>
- </array>
-
- <!-- The earliest hour of a day at which an event might be scheduled. -->
- <key>beginHour</key>
- <integer>8</integer>
-
- <!-- And the latest hour of a day (at which an event will be scheduled
- to begin!). -->
- <key>endHour</key>
- <integer>16</integer>
-
- <!-- The timezone in which the event is scheduled. (XXX Does this
- really work right?) -->
- <key>tzname</key>
- <string>America/Los_Angeles</string>
- </dict>
- </dict>
-
- <!-- Define how recurrences are created. -->
- <key>recurrenceDistribution</key>
- <dict>
-
- <!-- This distribution is pretty specialized. We have a fixed set of
- RRULEs defined for this distribution and pick each based on a
- weight. -->
- <key>type</key>
- <string>contrib.performance.stats.RecurrenceDistribution</string>
-
- <key>params</key>
- <dict>
- <!-- False to disable RRULEs -->
- <key>allowRecurrence</key>
- <true/>
-
- <!-- These are the weights for the specific set of RRULEs. -->
- <key>weights</key>
- <dict>
- <!-- Half of all events will be non-recurring -->
- <key>none</key>
- <integer>50</integer>
-
- <!-- Daily and weekly are pretty common -->
- <key>daily</key>
- <integer>25</integer>
- <key>weekly</key>
- <integer>25</integer>
-
- <!-- Monthly, yearly, daily & weekly limit not so common -->
- <key>monthly</key>
- <integer>0</integer>
- <key>yearly</key>
- <integer>0</integer>
- <key>dailylimit</key>
- <integer>0</integer>
- <key>weeklylimit</key>
- <integer>0</integer>
-
- <!-- Work days pretty common -->
- <key>workdays</key>
- <integer>0</integer>
- </dict>
- </dict>
- </dict>
- </dict>
- </dict>
-
- <!-- This profile invites some number of new attendees to new events. -->
- <dict>
- <key>class</key>
- <string>contrib.performance.loadtest.profiles.RealisticInviter</string>
-
- <key>params</key>
- <dict>
- <key>enabled</key>
- <true/>
-
- <!-- Define the frequency at which new invitations will be sent out. -->
- <key>sendInvitationDistribution</key>
- <dict>
- <key>type</key>
- <string>contrib.performance.stats.FixedDistribution</string>
- <key>params</key>
- <dict>
- <!-- interval (in seconds). -->
- <key>value</key>
- <integer>120</integer>
- </dict>
- </dict>
-
- <!-- Define the distribution of who will be invited to an event.
-
- When inviteeClumping is turned on each invitee is based on a sample of
- users "close to" the organizer based on account index. If the clumping
- is too "tight" for the requested number of attendees, then invites for
- those larger numbers will simply fail (the sim will report that situation).
-
- When inviteeClumping is off invitees will be sampled across an entire
- range of account indexes. In this case the distribution ought to be a
- UniformIntegerDistribution with min=0 and max set to the number of accounts.
- -->
- <key>inviteeDistribution</key>
- <dict>
- <key>type</key>
- <string>contrib.performance.stats.UniformIntegerDistribution</string>
- <key>params</key>
- <dict>
- <!-- The minimum value (inclusive) of the uniform distribution. -->
- <key>min</key>
- <integer>0</integer>
- <!-- The maximum value (exclusive) of the uniform distribution. -->
- <key>max</key>
- <integer>99</integer>
- </dict>
- </dict>
-
- <key>inviteeClumping</key>
- <true/>
-
- <!-- Define the distribution of how many attendees will be invited to an event.
-
- LogNormal is the best fit to observed data.
-
-
- For LogNormal "mode" is the peak, "mean" is the mean value. For invites,
- mode should typically be 1, and mean whatever matches the user behavior.
- Our typical mean is 6.
- -->
- <key>inviteeCountDistribution</key>
- <dict>
- <key>type</key>
- <string>contrib.performance.stats.FixedDistribution</string>
- <key>params</key>
- <dict>
- <!-- Number of attendees. -->
- <key>value</key>
- <integer>5</integer>
- </dict>
- </dict>
-
- <!-- Define how start times (DTSTART) for the randomly generated events
- will be selected. This is an example of a "Distribution" parameter. The value
- for most "Distribution" parameters are interchangeable and extensible. -->
- <key>eventStartDistribution</key>
- <dict>
-
- <!-- This distribution is pretty specialized. It produces timestamps
- in the near future, limited to certain days of the week and certain hours
- of the day. -->
- <key>type</key>
- <string>contrib.performance.stats.WorkDistribution</string>
-
- <key>params</key>
- <dict>
- <!-- These are the days of the week the distribution will use. -->
- <key>daysOfWeek</key>
- <array>
- <string>mon</string>
- <string>tue</string>
- <string>wed</string>
- <string>thu</string>
- <string>fri</string>
- </array>
-
- <!-- The earliest hour of a day at which an event might be scheduled. -->
- <key>beginHour</key>
- <integer>8</integer>
-
- <!-- And the latest hour of a day (at which an event will be scheduled
- to begin!). -->
- <key>endHour</key>
- <integer>16</integer>
-
- <!-- The timezone in which the event is scheduled. (XXX Does this
- really work right?) -->
- <key>tzname</key>
- <string>America/Los_Angeles</string>
- </dict>
- </dict>
-
- <!-- Define how recurrences are created. -->
- <key>recurrenceDistribution</key>
- <dict>
-
- <!-- This distribution is pretty specialized. We have a fixed set of
- RRULEs defined for this distribution and pick each based on a
- weight. -->
- <key>type</key>
- <string>contrib.performance.stats.RecurrenceDistribution</string>
-
- <key>params</key>
- <dict>
- <!-- False to disable RRULEs -->
- <key>allowRecurrence</key>
- <false/>
-
- <!-- These are the weights for the specific set of RRULEs. -->
- <key>weights</key>
- <dict>
- <!-- Half of all events will be non-recurring -->
- <key>none</key>
- <integer>50</integer>
-
- <!-- Daily and weekly are pretty common -->
- <key>daily</key>
- <integer>10</integer>
- <key>weekly</key>
- <integer>20</integer>
-
- <!-- Monthly, yearly, daily & weekly limit not so common -->
- <key>monthly</key>
- <integer>2</integer>
- <key>yearly</key>
- <integer>1</integer>
- <key>dailylimit</key>
- <integer>2</integer>
- <key>weeklylimit</key>
- <integer>5</integer>
-
- <!-- Work days pretty common -->
- <key>workdays</key>
- <integer>10</integer>
- </dict>
- </dict>
- </dict>
- </dict>
- </dict>
-
- <!-- This profile accepts invitations to events, handles cancels, and
- handles replies received. -->
- <dict>
- <key>class</key>
- <string>contrib.performance.loadtest.profiles.Accepter</string>
-
- <key>params</key>
- <dict>
- <key>enabled</key>
- <false/>
-
- <!-- Define how long to wait after seeing a new invitation before
- accepting it.
-
- For LogNormal "mode" is the peak, "median" is the 50% cummulative value
- (i.e., half of the user have accepted by that time).
- -->
- <key>acceptDelayDistribution</key>
- <dict>
- <key>type</key>
- <string>contrib.performance.stats.LogNormalDistribution</string>
- <key>params</key>
- <dict>
- <!-- mode - peak-->
- <key>mode</key>
- <integer>300</integer>
- <!-- median - 50% done-->
- <key>median</key>
- <integer>1800</integer>
- </dict>
- </dict>
- </dict>
- </dict>
-
- <!-- A task-creating profile, which will periodically create
- new tasks at a random time on a random calendar. -->
- <dict>
- <key>class</key>
- <string>contrib.performance.loadtest.profiles.Tasker</string>
-
- <key>params</key>
- <dict>
- <key>enabled</key>
- <false/>
-
- <!-- Define the interval (in seconds) at which this profile will use
- its client to create a new task. -->
- <key>interval</key>
- <integer>300</integer>
-
- <!-- Define how due times (DUE) for the randomly generated tasks
- will be selected. This is an example of a "Distribution" parameter. The value
- for most "Distribution" parameters are interchangeable and extensible. -->
- <key>taskDueDistribution</key>
- <dict>
-
- <!-- This distribution is pretty specialized. It produces timestamps
- in the near future, limited to certain days of the week and certain hours
- of the day. -->
- <key>type</key>
- <string>contrib.performance.stats.WorkDistribution</string>
-
- <key>params</key>
- <dict>
- <!-- These are the days of the week the distribution will use. -->
- <key>daysOfWeek</key>
- <array>
- <string>mon</string>
- <string>tue</string>
- <string>wed</string>
- <string>thu</string>
- <string>fri</string>
- </array>
-
- <!-- The earliest hour of a day at which an event might be scheduled. -->
- <key>beginHour</key>
- <integer>8</integer>
-
- <!-- And the latest hour of a day (at which an event will be scheduled
- to begin!). -->
- <key>endHour</key>
- <integer>16</integer>
-
- <!-- The timezone in which the event is scheduled. (XXX Does this
- really work right?) -->
- <key>tzname</key>
- <string>America/Los_Angeles</string>
- </dict>
- </dict>
- </dict>
- </dict>
-
- </array>
-
- <!-- Determine the frequency at which this client configuration will
- appear in the clients which are created by the load tester. -->
- <key>weight</key>
- <integer>1</integer>
- </dict>
- </array>
- </dict>
-</plist>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.macosforge.org/pipermail/calendarserver-changes/attachments/20160622/c1727188/attachment-0001.html>
More information about the calendarserver-changes
mailing list